Director of Data Science

Unified Women's Healthcare

About The Position

Lucina is a maternity analytics platform that equips health plans and providers with innovative, AI-enabled insights that allow care management teams to identify high-risk mothers sooner and more effectively. Acquired by Unified Women's Healthcare in 2020, with the backing of Ares Management and Atlas Capital Partners, Lucina addresses a critical gap in women's health, enhancing outcomes across the journey of pregnancy, especially for women of color for whom the maternity care crisis is most prominent. Lucina's technology is user-friendly and easy to implement into care management systems. By analyzing thousands of signals and risk factors, Lucina's platform has helped health plans to identify 98% of expecting mothers before delivery and 70% in the first trimester. Additionally, by integrating AI with in-person, tailored care plans, Lucina has helped health plans see a 10% reduction in preterm birth rates and NICU utilization. Beyond pregnancy, Lucina also recently launched a predictive postpartum tool aimed to support new mothers in the crucial 12 weeks after birth in which pregnancy-related complications have been reported to escalate. Lucina is also pioneering an Integrated Care Management model that embeds care coordinators directly within Unified's provider practices to deliver white-labeled, high-touch support. This approach leverages Unified's significant market presence to create seamless care experiences for members. Lucina Analytics is seeking a Director of Data Science to lead the design, development, and deployment of advanced analytics, machine learning models, and AI-enabled capabilities across the Unified Women’s Health ecosystem. This role is responsible for translating complex clinical and operational data into scalable, production-ready models that drive measurable improvements in outcomes, cost, and value-based care performance. The Director of Data Science will own Lucina’s modeling roadmap—driving maternity and women’s health use cases—and will partner closely with Analytics, Engineering, Clinical, Product, and Actuarial teams to ensure models are clinically credible, create operational value, and are financially defensible. This is a hands-on leadership role: equal parts strategy, technical depth, and execution.

Requirements

Advanced degree in Data Science, Statistics, Applied Mathematics, Computer Science, Epidemiology, Biostatistics, or related field (PhD a plus, not required).
8+ years of experience in data science, with demonstrated ownership of ML models in production.
Strong experience with: Predictive modeling and statistical methods
Python/R and modern ML libraries
Working with healthcare data (claims, EMR, or population health data)
Proven ability to translate initially ambiguous business or clinical questions into analytic solutions.
Experience partnering cross-functionally in complex, matrixed environments.

Nice To Haves

Healthcare analytics, population health, or value-based care experience.
Experience building or scaling ML models in regulated environments.
Familiarity with: Model explainability techniques (SHAP, LIME, etc.)
Clinical risk adjustment
ROI attribution or intervention evaluation frameworks
Experience with cloud-based analytics platforms and data warehouses.

Responsibilities

Lead the development of predictive, prescriptive, and evaluative models, including but not limited to: Prospective risk stratification and risk scoring
Clinical utilization and outcome prediction (e.g., NICU admission, preterm birth, C-section)
Intervention impact and ROI attribution causal modeling
Population segmentation and cohort identification
Design model architectures that leverage claims, EMR, SDoH, and external datasets.
Ensure models are explainable, transparent, and appropriate for clinical and operational decision-making.
Identify and implement AI-enabled opportunities across Lucina products, including: Automation of analytic workflows
Embedding model-driven insights into dashboards and platforms
Scalable feature engineering and model retraining approaches
Partner with Engineering to establish MLOps best practices (versioning, monitoring, retraining, model governance).
Evaluate emerging AI/ML tools and frameworks with a practical lens—focused on value, safety, and scalability in healthcare.
Own the full lifecycle of models: problem definition → development → validation → deployment → monitoring.
Collaborate with Data Engineering to ensure models are built on a robust, governed enterprise data foundation.
Define success metrics and ongoing performance monitoring for all deployed models.
Build and lead a high-performing Data Science team (initially small, with room to scale).
Serve as a strategic partner to: Analytics & Reporting teams (to operationalize insights)
Product & Client Delivery teams (to embed models into workflows)
Clinical leaders (to ensure clinical relevance and adoption)
Actuarial and Finance partners (to support value-based care and ROI narratives)
Contribute to enterprise analytics governance, standards, and prioritization.